Cryosystem for DC spark experiments: Construction and acceptance tests

Department of

Physics and Astronomy Uppsala University P.O. Box 516

SE – 751 20 Uppsala Sweden

Papers in the FREIA Report Series are published on internet in PDF- formats.

Download from http://uu.diva-portal.org

FREIA Report 2019/02 March 29, 2019

Cryosystem for DC spark experiments

DEPARTMENT OF PHYSICS AND ASTRONOMY UPPSALA UNIVERSITY

Construction and acceptance tests

J. Eriksson, M. Jacewicz, R. Ruber

Uppsala University, Uppsala, Sweden

Cryosystem for DC spark experiments Construction and acceptance tests

J. Ericsson, M. Jacewicz, R. Ruber

FREIA Laboratory

Uppsala, 29. March 2019

1

Contents

1 Introduction 2

2 Design 3

2.1 Cryocooler . . . . 4

2.2 Cryostat . . . . 5

2.3 Electrode system . . . . 7

2.4 Temperature control . . . . 7

2.5 Vacuum system . . . . 7

2.6 Insulation test . . . . 8

2.7 Control system . . . . 8

3 Initial commissioning of the system 10 3.1 Cool-down . . . 10

3.2 Temperature behaviour . . . 11

3.3 Gap distance measurements . . . 11

3.4 First test with HV system . . . 13

3.5 Summary . . . 13

2

Chapter 1

Introduction

Field emission and vacuum arcs are phenomena which limit the accelerating gradient of normal and super- conducting accelerator cavities. An extensive program of developing and testing 100 MV/m-range, normal- conducting, X-band accelerating structures has been carried out in the radiofrequency (rf) test stands at CERN. To complement these rf tests, a very high repetition rate, pulsed, direct current (dc) system with simple, large planar electrodes has been built and operated. Results from the pulsed dc system have confirmed that high-field behaviors in rf and pulsed dc are quite similar, validating the use of the pulsed dc system for in depth studies of the fundamental physics of high-fields, for example of material and surface science, and for development of technology for high-gradient accelerating structures.

In order to extend the understanding of high-field physics and material science to low temperatures a new

version of the pulsed dc system has been developed. This new cryogenic experimental infrastructure will

enable to make important and fundamental contributions to the understanding of high-gradient physics and

potentially finding new connections between the high-gradient normal-conducting and superconducting fields.

3

Chapter 2

Design

The setup consists of a cryostat, where the experiments take place, cooled down by a stand-alone cryocooler.

Temperature inside the cryostat is regulated by a temperature controller. Ultra high vacuum is maintained by a external pumping station and the system can be monitored by a PC with LabView control software.

Figure 2.1: Conceptual layout of the system.

Chapter 2: Design 4

2.1 Cryocooler

The selected cryocooler is a pulse tube type CRYOMECH PT415 cryo-refrigerator [1]. The helium compres- sor supplying the PT415 cold head is CPA111i tri-phases 380V-415V / 50Hz water cooled type compressor equipped with an inverter. The inverter allows for variation of the compressor frequency which modifies the cooler capacity and the electrical input power thus allowing for more flexible operation and wider temperature range. The nominal performance with max. 78Hz compressor frequency is 1,5W @ 4.2K available at second stage with simultaneously 40W @ 45K load at first stage. The frequency can be changed between 40 and 78Hz.

Figure 2.2: Top: System under vacuum. Bottom: Main parts of the cryocooler. From the left: Main panel

and controls of the compressor and cold head installed on a main vacuum flange of the cryostat.

Chapter 2: Design 5

2.2 Cryostat

Cryostat consists of a stainless steel vacuum chamber that surrounds the cold head of the cryocooler. Inside the vacuum chamber, each stage of the cryocooler is protected from the thermal radiation by a copper cylindrical shield. The vacuum chamber is equipped with several vacuum flange of the con-flat type to allow for various connection of the outside equipment in a flexible way.

Figure 2.3: Main parts of the cryostat. From the left: first stage radiation shield in place, insertion of the second stage radiation shield and insertion of the vacuum chamber.

In figure 2.4 we present CAD drawing of the cryostat with indicated main dimensions.

Chapter 2: Design 6

Figure 2.4: CAD drawing of the cryostat.

Chapter 2: Design 7

2.3 Electrode system

Figure 2.5: The electrode setup with instrumentation.

The large electrode setup is attached on the second stage of the cryocooler. The top electrode has mechanical contact via copper cylinder and a copper flange with the cold finger of the stage. The copper flange also acts as a support for tempera- ture instrumentation: resistor heaters and temperature sensors. All cables are ther- mally anchored both on the first and the second stage of the cryocooler. The elec- trodes are in turn pressed together with the help of two alumina rings and five stainless steel rods with strong springs (approximately 5kN/m) keeping the ten- sion, see figure 2.5. This provides a robust support with minimal material budget.

The electrode system can be assembled in a separate room and easily mounted on the second stage as there are no fast con- nection between the parts.

2.4 Temperature control

During the operation the temperature inside the cryostat is controlled with the help of 5 temperature sensors (the system can be easily extended to 8 sensors if needed) and two control loops equipped with electrical heaters. The sensors in use are carbon-ceramic cryogenic temperature sensors from Temati [2]. The sensors are calibrated in the range from 300K to 4K. The resistance of the sensor is measured in 4-wire configuration.

Each control loop is assign to one of the cryocooler stages and is monitored by a sensor mounted directly on that stage. Two other sensors are mounted on the respective radiation shield of each stage. The last sensor is positioned on the ceramic spacer between the electrodes. For the control we use LakeShore Model 336-3062 temperature controller with 8 inputs and 2 PID control loops providing power for the electrical heaters [3]. In the cabling operations we used established procedures as describe in [4] and [5].

2.5 Vacuum system

The vacuum system consist of an Agilent complete UHV pumping station on a mobile cart [6] with Agilent

TV 301 Navigator turbomolecular pump with Agilent XGS-600 Gauge Controller driving ConvecTorr Rough

Vacuum gauge and IMG-100 High Vacuum gauge. In the warm state the typical vacuum level measured at

the pump is around 5 × 10

mbar . Cooling of the cryostat results in additional cryo vacuum pumping by cold

surfaces and vacuum levels around 8 × 10

mbar were measured.

Chapter 2: Design 8

2.6 Insulation test

Figure 2.6: Insulation test results.

The global dialectric insulation of the setup was verified with Insulation Resistance Tester Megger MIT525 [7].

Two test were performed:

1. A timed insulation test (1 minute) 2. DAR test

In the insulation test 290 GΩ insulation resistance was mea- sured.

The DAR test resulted in value of 1.55 which indicates excel- lent insulation condition according to the table provided by the Megger company. The photograph of the tester during mea- surements in presented in figure 2.6. We observed no issues with setup insulation during commissioning test either.

2.7 Control system

The whole setup can be monitored and control from a dedicated

PC running Labview program. The following inputs are monitored and logged by the system:

• Readings of the temperature sensors

• Heater status

• All the parameters of the compressor

• Vacuum gauges

• Capacitance between the electrodes

The PC is connected to the internal network in Freia lab and can used remotely. Figure 2.7 presents the first version of the monitoring program.

1

DAR is defined as the ratio of insulation resistance at 1 minute divided by insulation resistance at 30 seconds.

Chapter 2: Design 9

Figure 2.7: The first version of the Labview control system. The main panel displays current status of all

parameters as well as the history of the measurements in a graphical form.

10

Chapter 3

Initial commissioning of the system

3.1 Cool-down

The cooling of the system starts when the cryocooler is turned on. During the operation several parameters of the compressor are monitored, most important being :

• Warning and Alarm State

• Coolant Temp In and Out

• Oil Temp

• Helium Temp

• Low and High Pressure

• Motor Current

The cool-down time can be defined in different ways. Perhaps most interesting is the time after which the spacer between electrodes has reached given temperature. The curves in figure 3.1 show the temperature changes inside the cryostat. One can see in the right pane that the stable temperature below 40K at the second stage was obtained after about 6h of operation. The temperatures continue to drop thereafter with however much slower rate. The radiation shield on the other hand takes much longer to reached that temperature, of the order of 48h. The more powerful first stage of the cooler (left pane in the figure) is able to cool down the radiation shield of the first stage much faster, of the order of 10h. We were able to cool down the setup down to 20K. Further lowering of the temperature should be possible by improving thermal insulation inside the cryostat. One should note that 20K is considered sufficient for the first breakdown rate tests.

Figure 3.1: Cooling of the system. Curves for the temperature sensors during cool-down. Left: sensors on the

first stage, blue located on the cold-finger, orange on the radiation shield. Right: sensors on the second stage,

blue located on the cold-finger, green on the spacer between electrodes and orange on the radiation shield.

Chapter 3: Initial commissioning of the system 11

3.2 Temperature behaviour

The temperature control system is able to precisely control the temperature inside the cryostat using two independent PID control outputs supplying 100 W and 50 W of heater power associated with temperature sensors mounted on the cold-finger of stage 1 and 2 of the cryocooler. In the PID loops the control output is calculated based on your temperature set-point and feedback from the control sensor. That system works very well when the HV discharge system is off. See figure 3.2 left. The temperature fluctuations are well below 0.5K.

Figure 3.2: Temperature readout stability in case wher HV system is turned off (left) and during pulsing with HV (right). The vertical axes are different for the two cases.

During operation of the HV discharge system the current changes quickly while ramping the HV and especially rapidly during breakdowns, see figure 3.3 for example pulses recorded during tests.

Figure 3.3: Voltage and current measured during regular pulse (left) and during breakdown (right).

This changes are causing spurious reading of the temperatures. See figure 3.2 right. We believe that the reading do not reflect the actual fluctuations of the temperature, since they appear both below and above the set temperature line. As one can see from the plot the average temperature during the operation remains constant and tracks the set value (80K in this case). All the cabling ,except for HV connections, was done using twisted-pair wires. The fluctuations are most likely caused by sensors picking up the noise at the electrode system. These fluctuation will be studied further and the readout should be improved by signal filtering.

3.3 Gap distance measurements

During cool down process the distance between electrodes will change. In order to estimate the changes in the gap we measure the capacitance of the large electrode system with a LCR meter. The measurement cannot be done during operation of the HV system and instead is performed before and after temperature change.

Below we show the measurement of the electrode system capacitance in the function of time and temperature,

during cool-down, see figure 3.4.

Chapter 3: Initial commissioning of the system 12

Figure 3.4: Recorded changes in temperature and capacitance during cool-down.

The capacitance decreases due to the fact that the thermal contraction of copper electrodes is larger than of the alumina spacer causing the gap between electrodes to increase (also the area of electrodes shrinks). Below (in figure 3.5 we present the results of the calculations of the capacitance changes in function of time based on the thermal expansion coefficients available in literature [8][4].

Figure 3.5: Calculated changes in capacitance and gap distance based on literature data.

The measured capacitance differs from the calculated one due to additional cable and stray capacitance.

Also the calculated distance is only an approximation since we do not have an exact values for temperature expansion coefficient of the used ceramic material.

We suggest the following procedure for determining the gap size at given temperature:

To estimate the change in gap distance we measured system’s capacitance at stable temperature conditions (in our case we used 301K and 80K) and use the known formula 3.1

D = 



C (3.1)

In the measurements we obtained: C

= 513 pF and C

= 345 pF.

This corresponds to gap change of +22.9 µm. Assuming that the 60 µm gap size of copper electrodes was

measured at stable room temperature of 293K we expect –3 µm change of the gap for 301K case. That gives

in total (60 + 22.9 − −3) = 79.9µm gap in the case of 80K.

Chapter 3: Initial commissioning of the system 13

3.4 First test with HV system

The first test of the system was performed with Hard Copper electrodes no. 25 from CERN. The HV system together with control program was delivered for the test by CERN

The test started in room temperature with HV pulsing at 2kHz repetition rate. The pulse length was 1µs.

The first breakdown occurred at 23 MV/m surface field. In total we delivered 10

pulses with 832 BDs and reached approximately 62 MV/m field. The comparison for the conditioning with previously tested electrodes is shown in figure 3.6.

Figure 3.6: Results of initial conditioning compared with previously tested ones.

Afterwards the system was cooled-down to 80K and conditioning was restarted from low electric field value.

This time the breakdowns started at about 40 MV/m and conditioning continued as expected.

Figure 3.7: Progress curve from conditioning of the electrodes at 80K.

3.5 Summary

After successful commissioning the cold DC system for breakdown studies is now fully operational. The first

HV tests were performed in room temperature and at 80K. The only observed issue is a electronic noise on

the temperature lines during pulsing. This effect can be mitigated with better shielding and filtering of the

signals. Presently the minimum operational temperature is 20K, which is quite acceptable for the needs of the

discharge studies.

BIBLIOGRAPHY 14

Bibliography

[1] Cryomech PT415. 4K Two-Stage Pulse Tube Cryocooler. url: https://www.cryomech.com/products/

pt415/ .

[2] CCS. Carbon Ceramic Cryogenic Thermometers. url: https : / / www . temati - uk . com / html / ccs _ sensors.html .

[3] Model 336-3062. Temperature Controller. url: https://shop.lakeshore.com/temperature-products/

temperature- controllers/model- 336- temperature- controller/336- temperature- controller- with-3062-card.html .

[4] Jack Ekin. Experimental Techniques for Low-Temperature Measurements. Oxford Scholarship Online. Ox- ford University Press, 2006. isbn: 9780198570547.

[5] Ch. Balle et al. Installation guide for LHC cryogenic Thermometers. EDMS Document No. 110748. LHC- ACR-IN, 2002.

[6] Agilent TPS. Turbo Pumping System. url: https : / / www . agilent . com / en / products / vacuum - technologies/high-vacuum-pumps/turbo-pumping-systems/tps-mobile .

[7] Megger MIT525. 5 kV diagnostic insulation resistance tester. url: https://us.megger.com/5- kv- insulation-resistance-tester-mit525 .

[8] N.J. Simon. “Cryogenic Properties of Inorganic Insulation Materials for ITER Magnets: A Review”. In:

(Dec. 1994). doi: 10.2172/761710.

BIBLIOGRAPHY